I spent some time today looking at various birdies on my Linrad
screen, in an attempt to identify the source of each. By far the worst
are a series of lines centred around 144.153MHz. Each line is actually
a triplet, and the centre lines of each triplet are spaced 1kHz apart.
The intensities fall off on either side of 144.153MHz, making the
whole spectrum from 144.145 up quite difficult to use. This
investigation was prompted by JT1UN operating on 144.150MHz,
fortunately midway beween two of the triplet lines, though Faraday one-
way propagation conspired to frustrate all my attempts to work Jeremy.
Pickup is through the antenna from something radiated inside the
shack. It appears that the SDR-IQ is the source of this interference,
which can be reduced somewhat by several clip-on ferrites on the USB
lead close to the SDR-IQ. The USB plug is too large to pass through
any of the ferrite rings I happen to have on hand, though I could
obviously buy some larger ones.
Does any other SDR-IQ user experience this kind of interference, and
if so is there an effective cure?
> Pickup is through the antenna from something radiated inside the > shack. It appears that the SDR-IQ is the source of this interference, > which can be reduced somewhat by several clip-on ferrites on the USB > lead close to the SDR-IQ. The USB plug is too large to pass through > any of the ferrite rings I happen to have on hand, though I could > obviously buy some larger ones.
I suggest you wrap your SDR-IQ carefully in aluminium foil and remove the outer plastic from the two cables entering the unit. They are both screened and you should make a good connection between each screen and the aluminium foil.
This way you will eliminate any voltage difference between the two cables leaving the wrapped unit and you will also eliminate any voltage between different points on the foil. All currents will flow inside the enclosure and you should easily reach 60 dB attenuation. More aluminium will improve until the limit that is set by the leakage through ths screen of the USB cable.
Make a small wire loop and connect to the 144 MHz input. It could be one turn with 20 mm diameter. When you place that loop against the various conductors in your setup you will easily see where currents of the interference frequencies flow. Once you know where the currents are it will be reasonably easy to do something about them (as long as they are inside your own house...)
> > Pickup is through the antenna from something radiated inside the
> > shack. It appears that the SDR-IQ is the source of this interference,
> > which can be reduced somewhat by several clip-on ferrites on the USB
> > lead close to the SDR-IQ. The USB plug is too large to pass through
> > any of the ferrite rings I happen to have on hand, though I could
> > obviously buy some larger ones.
>I suggest you wrap your SDR-IQ carefully in aluminium foil and
>remove the outer plastic from the two cables entering the unit.
>They are both screened and you should make a good connection
>between each screen and the aluminium foil.
>This way you will eliminate any voltage difference between the
>two cables leaving the wrapped unit and you will also eliminate
>any voltage between different points on the foil. All currents
>will flow inside the enclosure and you should easily reach 60 dB
>attenuation. More aluminium will improve until the limit that
>is set by the leakage through ths screen of the USB cable.
>Make a small wire loop and connect to the 144 MHz input. It
>could be one turn with 20 mm diameter. When you place that loop
>against the various conductors in your setup you will easily see
>where currents of the interference frequencies flow. Once you know
>where the currents are it will be reasonably easy to do something
>about them (as long as they are inside your own house...)
>73
>Leif / SM5BSZ
I enclosed my SDR-IQ inside a Hammond Diecast box, using a brass shield to mount DB-9 connector. The USB jack isn't particularly shielded and the cable shield is only connected to the computer ground. Computers often do not like their ground tied to radio equipment grounds, so beware! I measured my noise floor at -134 dBm using a BNC 50-ohm termination on the antenna connector. Very little evidence of birdies on the Spectravue display when terminated this way. But I see a fair amount when an antenna is connected. Perhaps use of clamp-on ferrites on the USB cable might be effective if there is leakage there.
http://www.kl7uw.com/SDR.htm
Thanks Leif and Ed for the advice.
I will explore these suggestions over the coming days.
Meanwhile the single most effective change has been to move the SDR-IQ
to a different (and lower) physical location.
Next change will be to replace the 2m long USB lead (or whatever its
length is)
with something much shorter, like 600mm.
Well, well, well - we learn new things every day!
The lid of the SDR-IQ is NOT electrically connected to the body of the
box.
Fortunately the circuit board and various cable connections ARE
grounded to the body of the box.
The lid is held in place with 8 small countersunk screws - M2.5?
The lid has been carefully painted, including the bevelled edges of
the screw holes,
thus preventing the screws from making electrical contact with the
lid.
The lid itself and the box are of course separated by 2 layers of
paint.
Just to make quite sure that no electrical connection occurs,
the screws themselves are also black anodized - another insulating
coating.
I removed the paint from the countersunk holes with a hand-held 90-
degree counterbore bit.
Then, lacking any other suitable screws or a more suitable tool,
I spent a happy and fiddly half-hour using a small flat file to remove
the anodizing from
the underside of the 8 screw-heads. I also lightly touched up the
threads to show some metal.
The result was the almost complete elimination of the offending
birdies - 99% anyway.
A bit of aluminium foil as suggested by Leif completely eliminated the
last traces.
>Well, well, well - we learn new things every day!
>The lid of the SDR-IQ is NOT electrically connected to the body of the
>box.
>Fortunately the circuit board and various cable connections ARE
>grounded to the body of the box.
>The lid is held in place with 8 small countersunk screws - M2.5?
>The lid has been carefully painted, including the bevelled edges of
>the screw holes,
>thus preventing the screws from making electrical contact with the
>lid.
>The lid itself and the box are of course separated by 2 layers of
>paint.
>Just to make quite sure that no electrical connection occurs,
>the screws themselves are also black anodized - another insulating
>coating.
>I removed the paint from the countersunk holes with a hand-held 90-
>degree counterbore bit.
>Then, lacking any other suitable screws or a more suitable tool,
>I spent a happy and fiddly half-hour using a small flat file to remove
>the anodizing from
>the underside of the 8 screw-heads. I also lightly touched up the
>threads to show some metal.
>The result was the almost complete elimination of the offending
>birdies - 99% anyway.
>A bit of aluminium foil as suggested by Leif completely eliminated the
>last traces.
>Every SDR-IQ owner needs to do this!
>73 Guy VK2KU
The original issue of the SDR-IQ was without enclosure. As already mentioned, I installed mine in a Hammond Diecast box. Isolating the receiver from outside interference was the main reason. Funny how often digital engineers miss the finer points of RF. Good fix for those with the factory enclosure.
You might want to try replacing ferrite beads FB4 and FB6 on the DB-9
connector with shorts (solder blobs). We have identified this as a
potential source of EMI. The beads on the USB connector should already
be 0 Ohm resistors. Some of the bare boards that were shipped without
enclosures might still have beads in locations FB1 and FB3. These can
also be replaced with shorts.
regards,
Pieter
On Oct 16, 11:23 pm, VK2KU <vk...@clearmail.com.au> wrote:
> Well, well, well - we learn new things every day!
> The lid of the SDR-IQ is NOT electrically connected to the body of the
> box.
> Fortunately the circuit board and various cable connections ARE
> grounded to the body of the box.
> The lid is held in place with 8 small countersunk screws - M2.5?
> The lid has been carefully painted, including the bevelled edges of
> the screw holes,
> thus preventing the screws from making electrical contact with the
> lid.
> The lid itself and the box are of course separated by 2 layers of
> paint.
> Just to make quite sure that no electrical connection occurs,
> the screws themselves are also black anodized - another insulating
> coating.
> I removed the paint from the countersunk holes with a hand-held 90-
> degree counterbore bit.
> Then, lacking any other suitable screws or a more suitable tool,
> I spent a happy and fiddly half-hour using a small flat file to remove
> the anodizing from
> the underside of the 8 screw-heads. I also lightly touched up the
> threads to show some metal.
> The result was the almost complete elimination of the offending
> birdies - 99% anyway.
> A bit of aluminium foil as suggested by Leif completely eliminated the
> last traces.
> I removed the paint from the countersunk holes with a hand-held 90- > degree counterbore bit. > Then, lacking any other suitable screws or a more suitable tool, > I spent a happy and fiddly half-hour using a small flat file to remove > the anodizing from > the underside of the 8 screw-heads. I also lightly touched up the > threads to show some metal.
How about making the lower edge of the top of the box and the lip of the lower part electrically connected instead?
I've not yet taken my SDR-IQ apart, so I don't know if this is feasible, but it seems less fiddly than filing the countersunk part of the screw heads...
I could clean up the lid easily, but cleaning up the box is another
thing.
I was worried about paint and other chips of stuff getting into the
fine circuit board.
I guess I could take the whole board out, but cleaning up the screws
was just easier.
Guy VK2KU
On Oct 19, 5:33 am, Morseman <david.ackr...@gmail.com> wrote:
> How about making the lower edge of the top of the box and the lip of the
> lower part electrically connected instead?
> I've not yet taken my SDR-IQ apart, so I don't know if this is feasible, but
> it seems less fiddly than filing the countersunk part of the screw heads...
> I could clean up the lid easily, but cleaning up the box is another > thing. > I was worried about paint and other chips of stuff getting into the > fine circuit board. > I guess I could take the whole board out, but cleaning up the screws > was just easier.
I lost my way back here because I have to admit that this forum is not one of the most active in my listings. ;-)
OK, so today I decided to improve the connection between the upper and lower parts of my SDR-IQ. I took apart the case of my SDR-IQ and soon realised that I could do a lot without any problems of scattering material over the PCB.
The first thing I did was to use a small file (I would have used a dremel but I could not remember where I had put it) to expose the metal around the various screw holes inside the upper box lid.
Then I looked at the lower casing. Most of the screw fixings are on the other side of large metal walls. So careful filing exposes metal around most of the screw holes.
The two screw holes at the front of the lower case do need a bit more care, so I put a stiff card under the lip and then filed the paint away for those holes as well. Removing the filings carefully from under the lip of the case. I also filed the two lugs that the two back screws mate with to increase the metal to metal contact.
Unfortunately, once the case is put back together there is no obvious way to confirm connections but I will look out for any VHF Birdies in future.
I did consider smearing some copper based connection cream onto the exposed panels. However, I do know that Aluminum and Copper don't get on well but as most Amateur Radio stations are either short lived, or replace connections regularly, this may not be a problem...
|
